Detection of Dense Citrus Fruits by Combining Coordinated Attention and Cross-Scale Connection with Weighted Feature Fusion
The accuracy detection of individual citrus fruits in a citrus orchard environments is one of the key steps in realizing precision agriculture applications such as yield estimation, fruit thinning, and mechanical harvesting. This study proposes an improved object detection YOLOv5 model to achieve ac...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-06-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/13/6600 |
| _version_ | 1797480877656113152 |
|---|---|
| author | Xiaoyu Liu Guo Li Wenkang Chen Binghao Liu Ming Chen Shenglian Lu |
| author_facet | Xiaoyu Liu Guo Li Wenkang Chen Binghao Liu Ming Chen Shenglian Lu |
| author_sort | Xiaoyu Liu |
| collection | DOAJ |
| description | The accuracy detection of individual citrus fruits in a citrus orchard environments is one of the key steps in realizing precision agriculture applications such as yield estimation, fruit thinning, and mechanical harvesting. This study proposes an improved object detection YOLOv5 model to achieve accurate the identification and counting of citrus fruits in an orchard environment. First, the latest visual attention mechanism coordinated attention module (CA) was inserted into an improved backbone network to focus on fruit-dense regions to recognize small target fruits. Second, an efficient two-way cross-scale connection and weighted feature fusion BiFPN in the neck network were used to replace the PANet multiscale feature fusion network, giving effective feature corresponding weights to fully fuse the high-level and bottom-level features. Finally, the varifocal loss function was used to calculate the model loss for better model training results. The results of the experiments on four varieties of citrus trees showed that our improved model proposed to this study could effectively identify dense small citrus fruits. Specifically, the recognized AP (average precision) reached 98.4%, and the average recognition time was 0.019 s per image. Compared with the original YOLOv5 (including deferent variants of n, s, m, l, and x), the increase in the average accuracy precision of the improved YOLOv5 ranged from 7.5% to 0.8% while maintaining similar average inference time. Four different citrus varieties were also tested to evaluate the generalization performance of the improved model. The method can be further used as a part in a vision system to provide technical support for the real-time and accurate detection of multiple fruit targets during mechanical picking in citrus orchards. |
| first_indexed | 2024-03-09T22:06:30Z |
| format | Article |
| id | doaj.art-5e9252c82f81400b80dbcaabde2f35be |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T22:06:30Z |
| publishDate | 2022-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-5e9252c82f81400b80dbcaabde2f35be2023-11-23T19:39:33ZengMDPI AGApplied Sciences2076-34172022-06-011213660010.3390/app12136600Detection of Dense Citrus Fruits by Combining Coordinated Attention and Cross-Scale Connection with Weighted Feature FusionXiaoyu Liu0Guo Li1Wenkang Chen2Binghao Liu3Ming Chen4Shenglian Lu5Guangxi Key Lab of Multisource Information Mining & Security, College of Computer Science & Engineering, Guangxi Normal University, Guilin 541004, ChinaGuangxi Key Lab of Multisource Information Mining & Security, College of Computer Science & Engineering, Guangxi Normal University, Guilin 541004, ChinaGuangxi Key Lab of Multisource Information Mining & Security, College of Computer Science & Engineering, Guangxi Normal University, Guilin 541004, ChinaGuangxi Citrus Breeding and Cultivation Engineering Technology Center, Guangxi Academy of Specialty Crops, Guilin 541004, ChinaGuangxi Key Lab of Multisource Information Mining & Security, College of Computer Science & Engineering, Guangxi Normal University, Guilin 541004, ChinaGuangxi Key Lab of Multisource Information Mining & Security, College of Computer Science & Engineering, Guangxi Normal University, Guilin 541004, ChinaThe accuracy detection of individual citrus fruits in a citrus orchard environments is one of the key steps in realizing precision agriculture applications such as yield estimation, fruit thinning, and mechanical harvesting. This study proposes an improved object detection YOLOv5 model to achieve accurate the identification and counting of citrus fruits in an orchard environment. First, the latest visual attention mechanism coordinated attention module (CA) was inserted into an improved backbone network to focus on fruit-dense regions to recognize small target fruits. Second, an efficient two-way cross-scale connection and weighted feature fusion BiFPN in the neck network were used to replace the PANet multiscale feature fusion network, giving effective feature corresponding weights to fully fuse the high-level and bottom-level features. Finally, the varifocal loss function was used to calculate the model loss for better model training results. The results of the experiments on four varieties of citrus trees showed that our improved model proposed to this study could effectively identify dense small citrus fruits. Specifically, the recognized AP (average precision) reached 98.4%, and the average recognition time was 0.019 s per image. Compared with the original YOLOv5 (including deferent variants of n, s, m, l, and x), the increase in the average accuracy precision of the improved YOLOv5 ranged from 7.5% to 0.8% while maintaining similar average inference time. Four different citrus varieties were also tested to evaluate the generalization performance of the improved model. The method can be further used as a part in a vision system to provide technical support for the real-time and accurate detection of multiple fruit targets during mechanical picking in citrus orchards.https://www.mdpi.com/2076-3417/12/13/6600computer visioncitrus detectionYOLOv5small objectsreal-time detection |
| spellingShingle | Xiaoyu Liu Guo Li Wenkang Chen Binghao Liu Ming Chen Shenglian Lu Detection of Dense Citrus Fruits by Combining Coordinated Attention and Cross-Scale Connection with Weighted Feature Fusion Applied Sciences computer vision citrus detection YOLOv5 small objects real-time detection |
| title | Detection of Dense Citrus Fruits by Combining Coordinated Attention and Cross-Scale Connection with Weighted Feature Fusion |
| title_full | Detection of Dense Citrus Fruits by Combining Coordinated Attention and Cross-Scale Connection with Weighted Feature Fusion |
| title_fullStr | Detection of Dense Citrus Fruits by Combining Coordinated Attention and Cross-Scale Connection with Weighted Feature Fusion |
| title_full_unstemmed | Detection of Dense Citrus Fruits by Combining Coordinated Attention and Cross-Scale Connection with Weighted Feature Fusion |
| title_short | Detection of Dense Citrus Fruits by Combining Coordinated Attention and Cross-Scale Connection with Weighted Feature Fusion |
| title_sort | detection of dense citrus fruits by combining coordinated attention and cross scale connection with weighted feature fusion |
| topic | computer vision citrus detection YOLOv5 small objects real-time detection |
| url | https://www.mdpi.com/2076-3417/12/13/6600 |
| work_keys_str_mv | AT xiaoyuliu detectionofdensecitrusfruitsbycombiningcoordinatedattentionandcrossscaleconnectionwithweightedfeaturefusion AT guoli detectionofdensecitrusfruitsbycombiningcoordinatedattentionandcrossscaleconnectionwithweightedfeaturefusion AT wenkangchen detectionofdensecitrusfruitsbycombiningcoordinatedattentionandcrossscaleconnectionwithweightedfeaturefusion AT binghaoliu detectionofdensecitrusfruitsbycombiningcoordinatedattentionandcrossscaleconnectionwithweightedfeaturefusion AT mingchen detectionofdensecitrusfruitsbycombiningcoordinatedattentionandcrossscaleconnectionwithweightedfeaturefusion AT shenglianlu detectionofdensecitrusfruitsbycombiningcoordinatedattentionandcrossscaleconnectionwithweightedfeaturefusion |